Making Sense of Mark to Market

January 01, 1994

U.S. bankers have been struggling lately to keep up with increased regulatory and congressional scrutiny of their activities. Now the accountants are getting in on the act. Within the last two years the Financial Accounting Standards Board, or FASB (pronounced FAZZBEE), the chief rule-making body for accountants, has approved several proposals that alter the way banks and other financial institutions make financial disclosures.

These new standards fall under the general category of market value accounting (MVA). MVA, also known as fair value accounting or marking to market, requires that an asset or liability be valued according to its market price, that is, the current price at which it could be sold. The call for MVA has gained momentum in recent years, in part because of the savings and loan (S&L) crisis, during which traditional bank accounting methods failed to reveal the huge unrealized losses imbedded in S&Ls' mortgage portfolios. Much of the S&L losses ultimately borne by taxpayers could have been averted, MVA proponents say, if bank accounting methods had reflected the current rather than historical value of assets and liabilities.

Market Value vs. Historical Cost Accounting

Historical cost accounting (HCA) is the accounting method traditionally used by most financial institutions. With HCA, assets and liabilities are reported at their contractual values, which may or may not equal their market values.1

A simple example will illustrate the point. Suppose a banker purchases a one-year fixed-rate security (or makes a lump-sum, one-year loan) with an interest rate of 10 percent, the current market interest rate. Suppose midway through the year, market interest rates rise to 12 percent. This rise in interest rates would reduce the price (or market value) at which the bank could sell the loan or security at mid-year.2 The market value of the instrument would fall because a buyer would need to be compensated for the difference between the interest rate on that instrument (10 percent) and the going market interest rate (12 percent). In contrast, by HCA, this asset would be valued at $1,000. The possibility that changes in market rates will cause changes in earnings or the value of portfolios of assets and liabilities (and hence capital) is called interest rate risk.

Of course, the market value of that instrument would also decline if the borrower paid back only half the principal: The banker loses the opportunity to invest and earn interest on the remaining $500 outstanding in addition to the loss of principal. This type of risk is called credit or default risk.

Market values, then, are based on expected cash flows and foregone investment opportunities. For an asset that is actively traded, like a government bond, the market price is the best estimation of the asset's true economic value. Full market value accounting would require all assets and liabilities to be valued in a similar fashion.

Is Half a Loaf Better Than No Loaf?

The use of MVA has been hotly debated for several decades. Proponents argue that MVA would reveal the closest approximation to the true economic value of a bank—its capital—and that it would help both shareholders and regulators better monitor a bank's financial condition. Opponents counter that MVA, while theoretically appealing, is impractical for financial institutions because the market values of most of their assets and liabilities are difficult, if not impossible, to measure accurately.

While the jury is still out on whether full MVA will ever be adopted, partial MVA is now a reality. The most recent FASB statement, FAS 115 (Accounting for Certain Investments in Debt and Equity Securities), went into effect January 1, and requires that all banks report certain portions of their investment portfolios at market value.3 Under FAS 115, a bank's securities portfolio will be divided into three parts: held to maturity, available for sale and trading securities.

The "held to maturity" category will include only those instruments that the bank has the "positive intent and financial ability" to hold to maturity; these assets will continue to be reported at historical cost. The "trading account" will consist of securities that are frequently bought and sold to generate profits on short-term price movements; these assets will be marked to market with any unrealized gains and losses reported in the income statement, as they have been for a number of years. All other securities will fall into the category "available for sale"; these will also be market, but their unrealized gains and losses will not be reflected in the income statement. Rather, they will form a separate component of shareholders' equity, and thus will affect the measured value of bank capital on the balance sheet.

According to proponents, one of FAS 115's major benefits is that bank capital will now reflect interest rate risk as well as credit risk. Credit risk is accounted for in a bank's allowance for loan losses, a contra item on the balance sheet. Until now, interest rate risk has not been reflected in banks' financial statements.4 Interest rate risk can be significant for banks, especially for those banks with large securities holdings.

One of the primary economic factors affecting the value of investment securities, or any bond holding, is interest rates. As illustrated in the example above, the inverse relationship between interest rates and (fixed-rate) bond prices implies that an increase in interest rates will depress the market value of banks' outstanding investment securities. If a bank were forced to sell a security before its maturity in this environment, the bank would incur a loss (as the security's selling price drops below its purchase price). With MVA, these interest rate risk effects are accounted for in the capital account: Unrealized gains would supplement capital while unrealized losses would be deducted from capital. A bank with unrealized losses would be forced to hold more capital (or reduce assets) to meet regulatory standards, which would better protect depositors and the Bank Insurance Fund in the event of insolvency.

Moreover, marking a portion of the investment portfolio to market will eliminate the incentive for an accounting abuse known as "gains trading." Banks that gains trade tend to sell those securities with unrealized gains, bolstering income and perhaps book capital, while keeping securities with unrealized losses on the books at historical cost, keeping book capital artificially high. Because unrealized gains and losses will be reflected in capital under FAS 115, the incentive for gains trading will essentially be eliminated. Gains trading will also be discouraged by penalties for those banks that attempt to move securities from the "held to maturity" category into one of the other two categories to take advantage of market value gains.

Most Bankers Would Rather Go Without

Most bankers oppose MVA, whether it is partial or full. While the objections are far-ranging, most have to do with feasibility and possible adverse effects on the banking industry. The principle objection to FAS 115 is that it ignores liabilities. Under partial MVA, measured capital is likely to be volatile as the value of assets fluctuates while the value of liabilities stays constant. Most banks hedge against interest rate risk by making adjustments on the liability side of the balance sheet—such as matching up fixed-rate assets with fixed-rate liabilities of equal duration—to offset fluctuations on the asset side. Under FAS 115, then, measured capital could be a misleading indicator of the actual amount of interest rate risk inherent in the banks' operations. Increased capital volatility could also raise the cost of capital for many banks.

The problem with full MVA is that most bank assets are difficult to measure at market value. Small commercial loans, for example, are not actively traded so an observable market price does not exist. A similar problem exists on the liability side, in that there is no agreed-upon method to determine the market value of nontraded liabilities like demand deposits. Methods of estimating market values for these nontraded assets and liabilities are likely to vary substantially across banks, making comparability a major problem. And because each market value estimate would have to be done on a case-by case basis, banks are likely to incur significant costs.

Thus, while MVA accounting methods are theoretically appealing and could potentially give bank regulators and depositors a clear picture of banks' financial health, there are some significant real world costs associated with MVA that will not be easy to measure or mitigate. If MVA is to make further inroads into official bank accounting, its supporters should demonstrate that its real world benefits will exceed its real world costs.

Thomas A. Pollmann provided research assistance.


  1. Technically, most assets and all liabilities are currently reported at amortized cost, meaning that, for example, as borrowers make principal and interest payments, the amount outstanding of the loan or security is reduced. The value of liabilities reflects accrued interest due that has not been paid. [back to text]
  2. In this case, the market value of this asset can be calculated with the following formula: MV = C1/(l + r) + ...+Cn/(1 + r)n +P/(l +r)n, where C=interest payment, r=discount or interest rate, n=time to maturity, e.g., five years, and P=principal payment. Initially the present value of this asset is $1000: the bank could take this amount and invest it at 10 percent to earn the same return as the loan or security. If market interest rates rise to 12 percent immediately, the market value of this instrument falls to $982.14, since ((C + P)/1.12) =$1100/1.12 = $982.14. [back to text]
  3. Technically, FAS 115 became effective with fiscal years beginning after December 15,1993; because most banks' fiscal years coincide with the calendar year, January 1, 1994, was D-day for most. [back to text]
  4. In addition to accounting for interest rate risk with market value accounting of the investment portfolio, interest rate risk will soon be incorporated into banks' risk-based capital requirements. [back to text]


Hempel, George H., and Donald G. Simonson. "The Case for Comprehensive Market-Value Reporting," Bank Accounting and Finance (Spring 1992), pp. 23-29.

Morris, David M. "The Case Against Market-Value Accounting: A Pragmatic View," Bank Accounting and Finance (Spring 1992), pp. 30-36.

Morris, Charles S., and Gordon H. Sellon, Jr. "Market Value Accounting for Banks: Pros and Cons," Federal Reserve Bank of Kansas City Economic Review (March/April 1991), pp. 5-19.

Pickering, C. J., and Randy Wade. "Accounting for Investments: Understanding the Implications of FASB 115," Independent Banker (August 1993).

Shaffer, Sherrill. "Marking Banks to Market," Federal Reserve Bank of Philadelphia Business Review (July/August 1992), pp. 13-22.

About the Author
Michelle Clark Neely
Michelle Clark Neely

Michelle Clark Neely is a visiting scholar with the Supervision Policy, Research and Analysis team at the Federal Reserve Bank of St. Louis.

Michelle Clark Neely
Michelle Clark Neely

Michelle Clark Neely is a visiting scholar with the Supervision Policy, Research and Analysis team at the Federal Reserve Bank of St. Louis.

Views expressed in Regional Economist are not necessarily those of the St. Louis Fed or Federal Reserve System.

For the latest insights from our economists and other St. Louis Fed experts, visit On the Economy and subscribe.

Email Us

Media questions

Back to Top